base-japanese-product-classification

This model is a fine-tuned version of tohoku-nlp/bert-base-japanese-v3 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9961

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	106	3.5780
No log	2.0	212	2.6684
No log	3.0	318	2.1910
No log	4.0	424	1.8626
2.5157	5.0	530	1.6483
2.5157	6.0	636	1.4929
2.5157	7.0	742	1.3580
2.5157	8.0	848	1.2658
2.5157	9.0	954	1.1877
0.9057	10.0	1060	1.1381
0.9057	11.0	1166	1.0945
0.9057	12.0	1272	1.0589
0.9057	13.0	1378	1.0071
0.9057	14.0	1484	1.0002
0.4676	15.0	1590	0.9647
0.4676	16.0	1696	0.9762
0.4676	17.0	1802	0.9129
0.4676	18.0	1908	0.9316
0.2485	19.0	2014	0.9119
0.2485	20.0	2120	0.8805
0.2485	21.0	2226	0.8808
0.2485	22.0	2332	0.8769
0.2485	23.0	2438	0.8701
0.1331	24.0	2544	0.8651
0.1331	25.0	2650	0.8621
0.1331	26.0	2756	0.8759
0.1331	27.0	2862	0.8702
0.1331	28.0	2968	0.8634
0.0748	29.0	3074	0.8778
0.0748	30.0	3180	0.8776
0.0748	31.0	3286	0.8486
0.0748	32.0	3392	0.8695
0.0748	33.0	3498	0.8479
0.0416	34.0	3604	0.8661
0.0416	35.0	3710	0.8731
0.0416	36.0	3816	0.8681
0.0416	37.0	3922	0.8942
0.0255	38.0	4028	0.8841
0.0255	39.0	4134	0.8842
0.0255	40.0	4240	0.8875
0.0255	41.0	4346	0.8760
0.0255	42.0	4452	0.8820
0.0166	43.0	4558	0.8975
0.0166	44.0	4664	0.8890
0.0166	45.0	4770	0.8795
0.0166	46.0	4876	0.8882
0.0166	47.0	4982	0.8950
0.0123	48.0	5088	0.8923
0.0123	49.0	5194	0.9018
0.0123	50.0	5300	0.8975
0.0123	51.0	5406	0.9078
0.0097	52.0	5512	0.9124
0.0097	53.0	5618	0.9250
0.0097	54.0	5724	0.9663
0.0097	55.0	5830	0.9651
0.0097	56.0	5936	0.9570
0.0078	57.0	6042	0.9530
0.0078	58.0	6148	0.9548
0.0078	59.0	6254	0.9490
0.0078	60.0	6360	0.9563
0.0078	61.0	6466	0.9614
0.0064	62.0	6572	0.9602
0.0064	63.0	6678	0.9614
0.0064	64.0	6784	0.9625
0.0064	65.0	6890	0.9587
0.0064	66.0	6996	0.9601
0.0055	67.0	7102	0.9664
0.0055	68.0	7208	0.9688
0.0055	69.0	7314	0.9725
0.0055	70.0	7420	0.9726
0.0047	71.0	7526	0.9693
0.0047	72.0	7632	0.9737
0.0047	73.0	7738	0.9720
0.0047	74.0	7844	0.9717
0.0047	75.0	7950	0.9683
0.0041	76.0	8056	0.9732
0.0041	77.0	8162	0.9740
0.0041	78.0	8268	0.9748
0.0041	79.0	8374	0.9789
0.0041	80.0	8480	0.9788
0.0036	81.0	8586	0.9788
0.0036	82.0	8692	0.9829
0.0036	83.0	8798	0.9842
0.0036	84.0	8904	0.9810
0.0032	85.0	9010	0.9862
0.0032	86.0	9116	0.9858
0.0032	87.0	9222	0.9881
0.0032	88.0	9328	0.9889
0.0032	89.0	9434	0.9902
0.003	90.0	9540	0.9909
0.003	91.0	9646	0.9927
0.003	92.0	9752	0.9926
0.003	93.0	9858	0.9942
0.003	94.0	9964	0.9949
0.0028	95.0	10070	0.9926
0.0028	96.0	10176	0.9938
0.0028	97.0	10282	0.9949
0.0028	98.0	10388	0.9960
0.0028	99.0	10494	0.9960
0.0027	100.0	10600	0.9961

Framework versions

• Transformers 4.39.1
• Pytorch 2.2.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2